diff options
Diffstat (limited to 'fs/xfs/xfs_itable.c')
| -rw-r--r-- | fs/xfs/xfs_itable.c | 1016 |
1 files changed, 397 insertions, 619 deletions
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index b93e14b86754..2aa37a4d2706 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -1,127 +1,182 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_types.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_log_format.h" +#include "xfs_trans_resv.h" #include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_alloc_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_btree.h" #include "xfs_ialloc.h" +#include "xfs_ialloc_btree.h" +#include "xfs_iwalk.h" #include "xfs_itable.h" #include "xfs_error.h" -#include "xfs_btree.h" -#include "xfs_trace.h" #include "xfs_icache.h" +#include "xfs_health.h" +#include "xfs_trans.h" -STATIC int -xfs_internal_inum( - xfs_mount_t *mp, - xfs_ino_t ino) +/* + * Bulk Stat + * ========= + * + * Use the inode walking functions to fill out struct xfs_bulkstat for every + * allocated inode, then pass the stat information to some externally provided + * iteration function. + */ + +struct xfs_bstat_chunk { + bulkstat_one_fmt_pf formatter; + struct xfs_ibulk *breq; + struct xfs_bulkstat *buf; +}; + +static inline bool +want_metadir_file( + struct xfs_inode *ip, + struct xfs_ibulk *breq) { - return (ino == mp->m_sb.sb_rbmino || ino == mp->m_sb.sb_rsumino || - (xfs_sb_version_hasquota(&mp->m_sb) && - xfs_is_quota_inode(&mp->m_sb, ino))); + return xfs_is_metadir_inode(ip) && (breq->flags & XFS_IBULK_METADIR); } /* - * Return stat information for one inode. - * Return 0 if ok, else errno. + * Fill out the bulkstat info for a single inode and report it somewhere. + * + * bc->breq->lastino is effectively the inode cursor as we walk through the + * filesystem. Therefore, we update it any time we need to move the cursor + * forward, regardless of whether or not we're sending any bstat information + * back to userspace. If the inode is internal metadata or, has been freed + * out from under us, we just simply keep going. + * + * However, if any other type of error happens we want to stop right where we + * are so that userspace will call back with exact number of the bad inode and + * we can send back an error code. + * + * Note that if the formatter tells us there's no space left in the buffer we + * move the cursor forward and abort the walk. */ -int +STATIC int xfs_bulkstat_one_int( - struct xfs_mount *mp, /* mount point for filesystem */ - xfs_ino_t ino, /* inode to get data for */ - void __user *buffer, /* buffer to place output in */ - int ubsize, /* size of buffer */ - bulkstat_one_fmt_pf formatter, /* formatter, copy to user */ - int *ubused, /* bytes used by me */ - int *stat) /* BULKSTAT_RV_... */ + struct xfs_mount *mp, + struct mnt_idmap *idmap, + struct xfs_trans *tp, + xfs_ino_t ino, + struct xfs_bstat_chunk *bc) { - struct xfs_icdinode *dic; /* dinode core info pointer */ + struct user_namespace *sb_userns = mp->m_super->s_user_ns; struct xfs_inode *ip; /* incore inode pointer */ - struct xfs_bstat *buf; /* return buffer */ - int error = 0; /* error value */ - - *stat = BULKSTAT_RV_NOTHING; - - if (!buffer || xfs_internal_inum(mp, ino)) - return XFS_ERROR(EINVAL); - - buf = kmem_alloc(sizeof(*buf), KM_SLEEP | KM_MAYFAIL); - if (!buf) - return XFS_ERROR(ENOMEM); - - error = xfs_iget(mp, NULL, ino, + struct inode *inode; + struct xfs_bulkstat *buf = bc->buf; + xfs_extnum_t nextents; + int error = -EINVAL; + vfsuid_t vfsuid; + vfsgid_t vfsgid; + + error = xfs_iget(mp, tp, ino, (XFS_IGET_DONTCACHE | XFS_IGET_UNTRUSTED), XFS_ILOCK_SHARED, &ip); - if (error) { - *stat = BULKSTAT_RV_NOTHING; - goto out_free; + if (error == -ENOENT || error == -EINVAL) + goto out_advance; + if (error) + goto out; + + /* Reload the incore unlinked list to avoid failure in inodegc. */ + if (xfs_inode_unlinked_incomplete(ip)) { + error = xfs_inode_reload_unlinked_bucket(tp, ip); + if (error) { + xfs_iunlock(ip, XFS_ILOCK_SHARED); + xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); + xfs_irele(ip); + return error; + } } ASSERT(ip != NULL); ASSERT(ip->i_imap.im_blkno != 0); + inode = VFS_I(ip); + vfsuid = i_uid_into_vfsuid(idmap, inode); + vfsgid = i_gid_into_vfsgid(idmap, inode); - dic = &ip->i_d; + /* + * If caller wants files from the metadata directories, push out the + * bare minimum information for enabling scrub. + */ + if (want_metadir_file(ip, bc->breq)) { + memset(buf, 0, sizeof(*buf)); + buf->bs_ino = ino; + buf->bs_gen = inode->i_generation; + buf->bs_mode = inode->i_mode & S_IFMT; + xfs_bulkstat_health(ip, buf); + buf->bs_version = XFS_BULKSTAT_VERSION_V5; + xfs_iunlock(ip, XFS_ILOCK_SHARED); + xfs_irele(ip); + + error = bc->formatter(bc->breq, buf); + if (!error || error == -ECANCELED) + goto out_advance; + goto out; + } + + /* If this is a private inode, don't leak its details to userspace. */ + if (IS_PRIVATE(inode) || xfs_is_sb_inum(mp, ino)) { + xfs_iunlock(ip, XFS_ILOCK_SHARED); + xfs_irele(ip); + error = -EINVAL; + goto out_advance; + } /* xfs_iget returns the following without needing * further change. */ - buf->bs_nlink = dic->di_nlink; - buf->bs_projid_lo = dic->di_projid_lo; - buf->bs_projid_hi = dic->di_projid_hi; + buf->bs_projectid = ip->i_projid; buf->bs_ino = ino; - buf->bs_mode = dic->di_mode; - buf->bs_uid = dic->di_uid; - buf->bs_gid = dic->di_gid; - buf->bs_size = dic->di_size; - buf->bs_atime.tv_sec = dic->di_atime.t_sec; - buf->bs_atime.tv_nsec = dic->di_atime.t_nsec; - buf->bs_mtime.tv_sec = dic->di_mtime.t_sec; - buf->bs_mtime.tv_nsec = dic->di_mtime.t_nsec; - buf->bs_ctime.tv_sec = dic->di_ctime.t_sec; - buf->bs_ctime.tv_nsec = dic->di_ctime.t_nsec; + buf->bs_uid = from_kuid(sb_userns, vfsuid_into_kuid(vfsuid)); + buf->bs_gid = from_kgid(sb_userns, vfsgid_into_kgid(vfsgid)); + buf->bs_size = ip->i_disk_size; + + buf->bs_nlink = inode->i_nlink; + buf->bs_atime = inode_get_atime_sec(inode); + buf->bs_atime_nsec = inode_get_atime_nsec(inode); + buf->bs_mtime = inode_get_mtime_sec(inode); + buf->bs_mtime_nsec = inode_get_mtime_nsec(inode); + buf->bs_ctime = inode_get_ctime_sec(inode); + buf->bs_ctime_nsec = inode_get_ctime_nsec(inode); + buf->bs_gen = inode->i_generation; + buf->bs_mode = inode->i_mode; + buf->bs_xflags = xfs_ip2xflags(ip); - buf->bs_extsize = dic->di_extsize << mp->m_sb.sb_blocklog; - buf->bs_extents = dic->di_nextents; - buf->bs_gen = dic->di_gen; - memset(buf->bs_pad, 0, sizeof(buf->bs_pad)); - buf->bs_dmevmask = dic->di_dmevmask; - buf->bs_dmstate = dic->di_dmstate; - buf->bs_aextents = dic->di_anextents; - buf->bs_forkoff = XFS_IFORK_BOFF(ip); - - switch (dic->di_format) { + buf->bs_extsize_blks = ip->i_extsize; + + nextents = xfs_ifork_nextents(&ip->i_df); + if (!(bc->breq->flags & XFS_IBULK_NREXT64)) + buf->bs_extents = min(nextents, XFS_MAX_EXTCNT_DATA_FORK_SMALL); + else + buf->bs_extents64 = nextents; + + xfs_bulkstat_health(ip, buf); + buf->bs_aextents = xfs_ifork_nextents(&ip->i_af); + buf->bs_forkoff = xfs_inode_fork_boff(ip); + buf->bs_version = XFS_BULKSTAT_VERSION_V5; + + if (xfs_has_v3inodes(mp)) { + buf->bs_btime = ip->i_crtime.tv_sec; + buf->bs_btime_nsec = ip->i_crtime.tv_nsec; + if (ip->i_diflags2 & XFS_DIFLAG2_COWEXTSIZE) + buf->bs_cowextsize_blks = ip->i_cowextsize; + } + + switch (ip->i_df.if_format) { case XFS_DINODE_FMT_DEV: - buf->bs_rdev = ip->i_df.if_u2.if_rdev; + buf->bs_rdev = sysv_encode_dev(inode->i_rdev); buf->bs_blksize = BLKDEV_IOSIZE; buf->bs_blocks = 0; break; case XFS_DINODE_FMT_LOCAL: - case XFS_DINODE_FMT_UUID: buf->bs_rdev = 0; buf->bs_blksize = mp->m_sb.sb_blocksize; buf->bs_blocks = 0; @@ -130,577 +185,300 @@ xfs_bulkstat_one_int( case XFS_DINODE_FMT_BTREE: buf->bs_rdev = 0; buf->bs_blksize = mp->m_sb.sb_blocksize; - buf->bs_blocks = dic->di_nblocks + ip->i_delayed_blks; + buf->bs_blocks = ip->i_nblocks + ip->i_delayed_blks; break; } xfs_iunlock(ip, XFS_ILOCK_SHARED); - IRELE(ip); - - error = formatter(buffer, ubsize, ubused, buf); + xfs_irele(ip); - if (!error) - *stat = BULKSTAT_RV_DIDONE; + error = bc->formatter(bc->breq, buf); + if (error == -ECANCELED) + goto out_advance; + if (error) + goto out; - out_free: - kmem_free(buf); +out_advance: + /* + * Advance the cursor to the inode that comes after the one we just + * looked at. We want the caller to move along if the bulkstat + * information was copied successfully; if we tried to grab the inode + * but it's no longer allocated; or if it's internal metadata. + */ + bc->breq->startino = ino + 1; +out: return error; } -/* Return 0 on success or positive error */ -STATIC int -xfs_bulkstat_one_fmt( - void __user *ubuffer, - int ubsize, - int *ubused, - const xfs_bstat_t *buffer) -{ - if (ubsize < sizeof(*buffer)) - return XFS_ERROR(ENOMEM); - if (copy_to_user(ubuffer, buffer, sizeof(*buffer))) - return XFS_ERROR(EFAULT); - if (ubused) - *ubused = sizeof(*buffer); - return 0; -} - +/* Bulkstat a single inode. */ int xfs_bulkstat_one( - xfs_mount_t *mp, /* mount point for filesystem */ - xfs_ino_t ino, /* inode number to get data for */ - void __user *buffer, /* buffer to place output in */ - int ubsize, /* size of buffer */ - int *ubused, /* bytes used by me */ - int *stat) /* BULKSTAT_RV_... */ + struct xfs_ibulk *breq, + bulkstat_one_fmt_pf formatter) { - return xfs_bulkstat_one_int(mp, ino, buffer, ubsize, - xfs_bulkstat_one_fmt, ubused, stat); -} + struct xfs_bstat_chunk bc = { + .formatter = formatter, + .breq = breq, + }; + struct xfs_trans *tp; + int error; + + if (breq->idmap != &nop_mnt_idmap) { + xfs_warn_ratelimited(breq->mp, + "bulkstat not supported inside of idmapped mounts."); + return -EINVAL; + } -#define XFS_BULKSTAT_UBLEFT(ubleft) ((ubleft) >= statstruct_size) + ASSERT(breq->icount == 1); -/* - * Return stat information in bulk (by-inode) for the filesystem. - */ -int /* error status */ -xfs_bulkstat( - xfs_mount_t *mp, /* mount point for filesystem */ - xfs_ino_t *lastinop, /* last inode returned */ - int *ubcountp, /* size of buffer/count returned */ - bulkstat_one_pf formatter, /* func that'd fill a single buf */ - size_t statstruct_size, /* sizeof struct filling */ - char __user *ubuffer, /* buffer with inode stats */ - int *done) /* 1 if there are more stats to get */ -{ - xfs_agblock_t agbno=0;/* allocation group block number */ - xfs_buf_t *agbp; /* agi header buffer */ - xfs_agi_t *agi; /* agi header data */ - xfs_agino_t agino; /* inode # in allocation group */ - xfs_agnumber_t agno; /* allocation group number */ - int chunkidx; /* current index into inode chunk */ - int clustidx; /* current index into inode cluster */ - xfs_btree_cur_t *cur; /* btree cursor for ialloc btree */ - int end_of_ag; /* set if we've seen the ag end */ - int error; /* error code */ - int fmterror;/* bulkstat formatter result */ - int i; /* loop index */ - int icount; /* count of inodes good in irbuf */ - size_t irbsize; /* size of irec buffer in bytes */ - xfs_ino_t ino; /* inode number (filesystem) */ - xfs_inobt_rec_incore_t *irbp; /* current irec buffer pointer */ - xfs_inobt_rec_incore_t *irbuf; /* start of irec buffer */ - xfs_inobt_rec_incore_t *irbufend; /* end of good irec buffer entries */ - xfs_ino_t lastino; /* last inode number returned */ - int nbcluster; /* # of blocks in a cluster */ - int nicluster; /* # of inodes in a cluster */ - int nimask; /* mask for inode clusters */ - int nirbuf; /* size of irbuf */ - int rval; /* return value error code */ - int tmp; /* result value from btree calls */ - int ubcount; /* size of user's buffer */ - int ubleft; /* bytes left in user's buffer */ - char __user *ubufp; /* pointer into user's buffer */ - int ubelem; /* spaces used in user's buffer */ - int ubused; /* bytes used by formatter */ + bc.buf = kzalloc(sizeof(struct xfs_bulkstat), + GFP_KERNEL | __GFP_RETRY_MAYFAIL); + if (!bc.buf) + return -ENOMEM; /* - * Get the last inode value, see if there's nothing to do. + * Grab an empty transaction so that we can use its recursive buffer + * locking abilities to detect cycles in the inobt without deadlocking. */ - ino = (xfs_ino_t)*lastinop; - lastino = ino; - agno = XFS_INO_TO_AGNO(mp, ino); - agino = XFS_INO_TO_AGINO(mp, ino); - if (agno >= mp->m_sb.sb_agcount || - ino != XFS_AGINO_TO_INO(mp, agno, agino)) { - *done = 1; - *ubcountp = 0; - return 0; - } - if (!ubcountp || *ubcountp <= 0) { - return EINVAL; - } - ubcount = *ubcountp; /* statstruct's */ - ubleft = ubcount * statstruct_size; /* bytes */ - *ubcountp = ubelem = 0; - *done = 0; - fmterror = 0; - ubufp = ubuffer; - nicluster = mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(mp) ? - mp->m_sb.sb_inopblock : - (XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog); - nimask = ~(nicluster - 1); - nbcluster = nicluster >> mp->m_sb.sb_inopblog; - irbuf = kmem_zalloc_greedy(&irbsize, PAGE_SIZE, PAGE_SIZE * 4); - if (!irbuf) - return ENOMEM; - - nirbuf = irbsize / sizeof(*irbuf); + tp = xfs_trans_alloc_empty(breq->mp); + error = xfs_bulkstat_one_int(breq->mp, breq->idmap, tp, + breq->startino, &bc); + xfs_trans_cancel(tp); + kfree(bc.buf); /* - * Loop over the allocation groups, starting from the last - * inode returned; 0 means start of the allocation group. - */ - rval = 0; - while (XFS_BULKSTAT_UBLEFT(ubleft) && agno < mp->m_sb.sb_agcount) { - cond_resched(); - error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp); - if (error) { - /* - * Skip this allocation group and go to the next one. - */ - agno++; - agino = 0; - continue; - } - agi = XFS_BUF_TO_AGI(agbp); - /* - * Allocate and initialize a btree cursor for ialloc btree. - */ - cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno); - irbp = irbuf; - irbufend = irbuf + nirbuf; - end_of_ag = 0; - /* - * If we're returning in the middle of an allocation group, - * we need to get the remainder of the chunk we're in. - */ - if (agino > 0) { - xfs_inobt_rec_incore_t r; - - /* - * Lookup the inode chunk that this inode lives in. - */ - error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, - &tmp); - if (!error && /* no I/O error */ - tmp && /* lookup succeeded */ - /* got the record, should always work */ - !(error = xfs_inobt_get_rec(cur, &r, &i)) && - i == 1 && - /* this is the right chunk */ - agino < r.ir_startino + XFS_INODES_PER_CHUNK && - /* lastino was not last in chunk */ - (chunkidx = agino - r.ir_startino + 1) < - XFS_INODES_PER_CHUNK && - /* there are some left allocated */ - xfs_inobt_maskn(chunkidx, - XFS_INODES_PER_CHUNK - chunkidx) & - ~r.ir_free) { - /* - * Grab the chunk record. Mark all the - * uninteresting inodes (because they're - * before our start point) free. - */ - for (i = 0; i < chunkidx; i++) { - if (XFS_INOBT_MASK(i) & ~r.ir_free) - r.ir_freecount++; - } - r.ir_free |= xfs_inobt_maskn(0, chunkidx); - irbp->ir_startino = r.ir_startino; - irbp->ir_freecount = r.ir_freecount; - irbp->ir_free = r.ir_free; - irbp++; - agino = r.ir_startino + XFS_INODES_PER_CHUNK; - icount = XFS_INODES_PER_CHUNK - r.ir_freecount; - } else { - /* - * If any of those tests failed, bump the - * inode number (just in case). - */ - agino++; - icount = 0; - } - /* - * In any case, increment to the next record. - */ - if (!error) - error = xfs_btree_increment(cur, 0, &tmp); - } else { - /* - * Start of ag. Lookup the first inode chunk. - */ - error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &tmp); - icount = 0; - } - /* - * Loop through inode btree records in this ag, - * until we run out of inodes or space in the buffer. - */ - while (irbp < irbufend && icount < ubcount) { - xfs_inobt_rec_incore_t r; - - /* - * Loop as long as we're unable to read the - * inode btree. - */ - while (error) { - agino += XFS_INODES_PER_CHUNK; - if (XFS_AGINO_TO_AGBNO(mp, agino) >= - be32_to_cpu(agi->agi_length)) - break; - error = xfs_inobt_lookup(cur, agino, - XFS_LOOKUP_GE, &tmp); - cond_resched(); - } - /* - * If ran off the end of the ag either with an error, - * or the normal way, set end and stop collecting. - */ - if (error) { - end_of_ag = 1; - break; - } - - error = xfs_inobt_get_rec(cur, &r, &i); - if (error || i == 0) { - end_of_ag = 1; - break; - } - - /* - * If this chunk has any allocated inodes, save it. - * Also start read-ahead now for this chunk. - */ - if (r.ir_freecount < XFS_INODES_PER_CHUNK) { - struct blk_plug plug; - /* - * Loop over all clusters in the next chunk. - * Do a readahead if there are any allocated - * inodes in that cluster. - */ - blk_start_plug(&plug); - agbno = XFS_AGINO_TO_AGBNO(mp, r.ir_startino); - for (chunkidx = 0; - chunkidx < XFS_INODES_PER_CHUNK; - chunkidx += nicluster, - agbno += nbcluster) { - if (xfs_inobt_maskn(chunkidx, nicluster) - & ~r.ir_free) - xfs_btree_reada_bufs(mp, agno, - agbno, nbcluster, - &xfs_inode_buf_ops); - } - blk_finish_plug(&plug); - irbp->ir_startino = r.ir_startino; - irbp->ir_freecount = r.ir_freecount; - irbp->ir_free = r.ir_free; - irbp++; - icount += XFS_INODES_PER_CHUNK - r.ir_freecount; - } - /* - * Set agino to after this chunk and bump the cursor. - */ - agino = r.ir_startino + XFS_INODES_PER_CHUNK; - error = xfs_btree_increment(cur, 0, &tmp); - cond_resched(); - } - /* - * Drop the btree buffers and the agi buffer. - * We can't hold any of the locks these represent - * when calling iget. - */ - xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); - xfs_buf_relse(agbp); - /* - * Now format all the good inodes into the user's buffer. - */ - irbufend = irbp; - for (irbp = irbuf; - irbp < irbufend && XFS_BULKSTAT_UBLEFT(ubleft); irbp++) { - /* - * Now process this chunk of inodes. - */ - for (agino = irbp->ir_startino, chunkidx = clustidx = 0; - XFS_BULKSTAT_UBLEFT(ubleft) && - irbp->ir_freecount < XFS_INODES_PER_CHUNK; - chunkidx++, clustidx++, agino++) { - ASSERT(chunkidx < XFS_INODES_PER_CHUNK); - - ino = XFS_AGINO_TO_INO(mp, agno, agino); - /* - * Skip if this inode is free. - */ - if (XFS_INOBT_MASK(chunkidx) & irbp->ir_free) { - lastino = ino; - continue; - } - /* - * Count used inodes as free so we can tell - * when the chunk is used up. - */ - irbp->ir_freecount++; - - /* - * Get the inode and fill in a single buffer. - */ - ubused = statstruct_size; - error = formatter(mp, ino, ubufp, ubleft, - &ubused, &fmterror); - if (fmterror == BULKSTAT_RV_NOTHING) { - if (error && error != ENOENT && - error != EINVAL) { - ubleft = 0; - rval = error; - break; - } - lastino = ino; - continue; - } - if (fmterror == BULKSTAT_RV_GIVEUP) { - ubleft = 0; - ASSERT(error); - rval = error; - break; - } - if (ubufp) - ubufp += ubused; - ubleft -= ubused; - ubelem++; - lastino = ino; - } - - cond_resched(); - } - /* - * Set up for the next loop iteration. - */ - if (XFS_BULKSTAT_UBLEFT(ubleft)) { - if (end_of_ag) { - agno++; - agino = 0; - } else - agino = XFS_INO_TO_AGINO(mp, lastino); - } else - break; - } - /* - * Done, we're either out of filesystem or space to put the data. + * If we reported one inode to userspace then we abort because we hit + * the end of the buffer. Don't leak that back to userspace. */ - kmem_free_large(irbuf); - *ubcountp = ubelem; - /* - * Found some inodes, return them now and return the error next time. - */ - if (ubelem) - rval = 0; - if (agno >= mp->m_sb.sb_agcount) { - /* - * If we ran out of filesystem, mark lastino as off - * the end of the filesystem, so the next call - * will return immediately. - */ - *lastinop = (xfs_ino_t)XFS_AGINO_TO_INO(mp, agno, 0); - *done = 1; - } else - *lastinop = (xfs_ino_t)lastino; - - return rval; + if (error == -ECANCELED) + error = 0; + + return error; +} + +static int +xfs_bulkstat_iwalk( + struct xfs_mount *mp, + struct xfs_trans *tp, + xfs_ino_t ino, + void *data) +{ + struct xfs_bstat_chunk *bc = data; + int error; + + error = xfs_bulkstat_one_int(mp, bc->breq->idmap, tp, ino, data); + /* bulkstat just skips over missing inodes */ + if (error == -ENOENT || error == -EINVAL) + return 0; + return error; } /* - * Return stat information in bulk (by-inode) for the filesystem. - * Special case for non-sequential one inode bulkstat. + * Check the incoming lastino parameter. + * + * We allow any inode value that could map to physical space inside the + * filesystem because if there are no inodes there, bulkstat moves on to the + * next chunk. In other words, the magic agino value of zero takes us to the + * first chunk in the AG, and an agino value past the end of the AG takes us to + * the first chunk in the next AG. + * + * Therefore we can end early if the requested inode is beyond the end of the + * filesystem or doesn't map properly. */ -int /* error status */ -xfs_bulkstat_single( - xfs_mount_t *mp, /* mount point for filesystem */ - xfs_ino_t *lastinop, /* inode to return */ - char __user *buffer, /* buffer with inode stats */ - int *done) /* 1 if there are more stats to get */ +static inline bool +xfs_bulkstat_already_done( + struct xfs_mount *mp, + xfs_ino_t startino) +{ + xfs_agnumber_t agno = XFS_INO_TO_AGNO(mp, startino); + xfs_agino_t agino = XFS_INO_TO_AGINO(mp, startino); + + return agno >= mp->m_sb.sb_agcount || + startino != XFS_AGINO_TO_INO(mp, agno, agino); +} + +/* Return stat information in bulk (by-inode) for the filesystem. */ +int +xfs_bulkstat( + struct xfs_ibulk *breq, + bulkstat_one_fmt_pf formatter) { - int count; /* count value for bulkstat call */ - int error; /* return value */ - xfs_ino_t ino; /* filesystem inode number */ - int res; /* result from bs1 */ + struct xfs_bstat_chunk bc = { + .formatter = formatter, + .breq = breq, + }; + struct xfs_trans *tp; + int error; + + if (breq->idmap != &nop_mnt_idmap) { + xfs_warn_ratelimited(breq->mp, + "bulkstat not supported inside of idmapped mounts."); + return -EINVAL; + } + if (xfs_bulkstat_already_done(breq->mp, breq->startino)) + return 0; + + bc.buf = kzalloc(sizeof(struct xfs_bulkstat), + GFP_KERNEL | __GFP_RETRY_MAYFAIL); + if (!bc.buf) + return -ENOMEM; /* - * note that requesting valid inode numbers which are not allocated - * to inodes will most likely cause xfs_imap_to_bp to generate warning - * messages about bad magic numbers. This is ok. The fact that - * the inode isn't actually an inode is handled by the - * error check below. Done this way to make the usual case faster - * at the expense of the error case. + * Grab an empty transaction so that we can use its recursive buffer + * locking abilities to detect cycles in the inobt without deadlocking. */ + tp = xfs_trans_alloc_empty(breq->mp); + error = xfs_iwalk(breq->mp, tp, breq->startino, breq->iwalk_flags, + xfs_bulkstat_iwalk, breq->icount, &bc); + xfs_trans_cancel(tp); + kfree(bc.buf); - ino = (xfs_ino_t)*lastinop; - error = xfs_bulkstat_one(mp, ino, buffer, sizeof(xfs_bstat_t), 0, &res); - if (error) { - /* - * Special case way failed, do it the "long" way - * to see if that works. - */ - (*lastinop)--; - count = 1; - if (xfs_bulkstat(mp, lastinop, &count, xfs_bulkstat_one, - sizeof(xfs_bstat_t), buffer, done)) - return error; - if (count == 0 || (xfs_ino_t)*lastinop != ino) - return error == EFSCORRUPTED ? - XFS_ERROR(EINVAL) : error; - else - return 0; - } - *done = 0; - return 0; + /* + * We found some inodes, so clear the error status and return them. + * The lastino pointer will point directly at the inode that triggered + * any error that occurred, so on the next call the error will be + * triggered again and propagated to userspace as there will be no + * formatted inodes in the buffer. + */ + if (breq->ocount > 0) + error = 0; + + return error; } -int -xfs_inumbers_fmt( - void __user *ubuffer, /* buffer to write to */ - const xfs_inogrp_t *buffer, /* buffer to read from */ - long count, /* # of elements to read */ - long *written) /* # of bytes written */ +/* Convert bulkstat (v5) to bstat (v1). */ +void +xfs_bulkstat_to_bstat( + struct xfs_mount *mp, + struct xfs_bstat *bs1, + const struct xfs_bulkstat *bstat) { - if (copy_to_user(ubuffer, buffer, count * sizeof(*buffer))) - return -EFAULT; - *written = count * sizeof(*buffer); - return 0; + /* memset is needed here because of padding holes in the structure. */ + memset(bs1, 0, sizeof(struct xfs_bstat)); + bs1->bs_ino = bstat->bs_ino; + bs1->bs_mode = bstat->bs_mode; + bs1->bs_nlink = bstat->bs_nlink; + bs1->bs_uid = bstat->bs_uid; + bs1->bs_gid = bstat->bs_gid; + bs1->bs_rdev = bstat->bs_rdev; + bs1->bs_blksize = bstat->bs_blksize; + bs1->bs_size = bstat->bs_size; + bs1->bs_atime.tv_sec = bstat->bs_atime; + bs1->bs_mtime.tv_sec = bstat->bs_mtime; + bs1->bs_ctime.tv_sec = bstat->bs_ctime; + bs1->bs_atime.tv_nsec = bstat->bs_atime_nsec; + bs1->bs_mtime.tv_nsec = bstat->bs_mtime_nsec; + bs1->bs_ctime.tv_nsec = bstat->bs_ctime_nsec; + bs1->bs_blocks = bstat->bs_blocks; + bs1->bs_xflags = bstat->bs_xflags; + bs1->bs_extsize = XFS_FSB_TO_B(mp, bstat->bs_extsize_blks); + bs1->bs_extents = bstat->bs_extents; + bs1->bs_gen = bstat->bs_gen; + bs1->bs_projid_lo = bstat->bs_projectid & 0xFFFF; + bs1->bs_forkoff = bstat->bs_forkoff; + bs1->bs_projid_hi = bstat->bs_projectid >> 16; + bs1->bs_sick = bstat->bs_sick; + bs1->bs_checked = bstat->bs_checked; + bs1->bs_cowextsize = XFS_FSB_TO_B(mp, bstat->bs_cowextsize_blks); + bs1->bs_dmevmask = 0; + bs1->bs_dmstate = 0; + bs1->bs_aextents = bstat->bs_aextents; +} + +struct xfs_inumbers_chunk { + inumbers_fmt_pf formatter; + struct xfs_ibulk *breq; +}; + +/* + * INUMBERS + * ======== + * This is how we export inode btree records to userspace, so that XFS tools + * can figure out where inodes are allocated. + */ + +/* + * Format the inode group structure and report it somewhere. + * + * Similar to xfs_bulkstat_one_int, lastino is the inode cursor as we walk + * through the filesystem so we move it forward unless there was a runtime + * error. If the formatter tells us the buffer is now full we also move the + * cursor forward and abort the walk. + */ +STATIC int +xfs_inumbers_walk( + struct xfs_mount *mp, + struct xfs_trans *tp, + xfs_agnumber_t agno, + const struct xfs_inobt_rec_incore *irec, + void *data) +{ + struct xfs_inumbers inogrp = { + .xi_startino = XFS_AGINO_TO_INO(mp, agno, irec->ir_startino), + .xi_alloccount = irec->ir_count - irec->ir_freecount, + .xi_allocmask = ~irec->ir_free, + .xi_version = XFS_INUMBERS_VERSION_V5, + }; + struct xfs_inumbers_chunk *ic = data; + int error; + + error = ic->formatter(ic->breq, &inogrp); + if (error && error != -ECANCELED) + return error; + + ic->breq->startino = XFS_AGINO_TO_INO(mp, agno, irec->ir_startino) + + XFS_INODES_PER_CHUNK; + return error; } /* * Return inode number table for the filesystem. */ -int /* error status */ +int xfs_inumbers( - xfs_mount_t *mp, /* mount point for filesystem */ - xfs_ino_t *lastino, /* last inode returned */ - int *count, /* size of buffer/count returned */ - void __user *ubuffer,/* buffer with inode descriptions */ - inumbers_fmt_pf formatter) + struct xfs_ibulk *breq, + inumbers_fmt_pf formatter) { - xfs_buf_t *agbp; - xfs_agino_t agino; - xfs_agnumber_t agno; - int bcount; - xfs_inogrp_t *buffer; - int bufidx; - xfs_btree_cur_t *cur; - int error; - xfs_inobt_rec_incore_t r; - int i; - xfs_ino_t ino; - int left; - int tmp; - - ino = (xfs_ino_t)*lastino; - agno = XFS_INO_TO_AGNO(mp, ino); - agino = XFS_INO_TO_AGINO(mp, ino); - left = *count; - *count = 0; - bcount = MIN(left, (int)(PAGE_SIZE / sizeof(*buffer))); - buffer = kmem_alloc(bcount * sizeof(*buffer), KM_SLEEP); - error = bufidx = 0; - cur = NULL; - agbp = NULL; - while (left > 0 && agno < mp->m_sb.sb_agcount) { - if (agbp == NULL) { - error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp); - if (error) { - /* - * If we can't read the AGI of this ag, - * then just skip to the next one. - */ - ASSERT(cur == NULL); - agbp = NULL; - agno++; - agino = 0; - continue; - } - cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno); - error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_GE, - &tmp); - if (error) { - xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); - cur = NULL; - xfs_buf_relse(agbp); - agbp = NULL; - /* - * Move up the last inode in the current - * chunk. The lookup_ge will always get - * us the first inode in the next chunk. - */ - agino += XFS_INODES_PER_CHUNK - 1; - continue; - } - } - error = xfs_inobt_get_rec(cur, &r, &i); - if (error || i == 0) { - xfs_buf_relse(agbp); - agbp = NULL; - xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); - cur = NULL; - agno++; - agino = 0; - continue; - } - agino = r.ir_startino + XFS_INODES_PER_CHUNK - 1; - buffer[bufidx].xi_startino = - XFS_AGINO_TO_INO(mp, agno, r.ir_startino); - buffer[bufidx].xi_alloccount = - XFS_INODES_PER_CHUNK - r.ir_freecount; - buffer[bufidx].xi_allocmask = ~r.ir_free; - bufidx++; - left--; - if (bufidx == bcount) { - long written; - if (formatter(ubuffer, buffer, bufidx, &written)) { - error = XFS_ERROR(EFAULT); - break; - } - ubuffer += written; - *count += bufidx; - bufidx = 0; - } - if (left) { - error = xfs_btree_increment(cur, 0, &tmp); - if (error) { - xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); - cur = NULL; - xfs_buf_relse(agbp); - agbp = NULL; - /* - * The agino value has already been bumped. - * Just try to skip up to it. - */ - agino += XFS_INODES_PER_CHUNK; - continue; - } - } - } - if (!error) { - if (bufidx) { - long written; - if (formatter(ubuffer, buffer, bufidx, &written)) - error = XFS_ERROR(EFAULT); - else - *count += bufidx; - } - *lastino = XFS_AGINO_TO_INO(mp, agno, agino); - } - kmem_free(buffer); - if (cur) - xfs_btree_del_cursor(cur, (error ? XFS_BTREE_ERROR : - XFS_BTREE_NOERROR)); - if (agbp) - xfs_buf_relse(agbp); + struct xfs_inumbers_chunk ic = { + .formatter = formatter, + .breq = breq, + }; + struct xfs_trans *tp; + int error = 0; + + if (xfs_bulkstat_already_done(breq->mp, breq->startino)) + return 0; + + /* + * Grab an empty transaction so that we can use its recursive buffer + * locking abilities to detect cycles in the inobt without deadlocking. + */ + tp = xfs_trans_alloc_empty(breq->mp); + error = xfs_inobt_walk(breq->mp, tp, breq->startino, breq->iwalk_flags, + xfs_inumbers_walk, breq->icount, &ic); + xfs_trans_cancel(tp); + + /* + * We found some inode groups, so clear the error status and return + * them. The lastino pointer will point directly at the inode that + * triggered any error that occurred, so on the next call the error + * will be triggered again and propagated to userspace as there will be + * no formatted inode groups in the buffer. + */ + if (breq->ocount > 0) + error = 0; + return error; } + +/* Convert an inumbers (v5) struct to a inogrp (v1) struct. */ +void +xfs_inumbers_to_inogrp( + struct xfs_inogrp *ig1, + const struct xfs_inumbers *ig) +{ + /* memset is needed here because of padding holes in the structure. */ + memset(ig1, 0, sizeof(struct xfs_inogrp)); + ig1->xi_startino = ig->xi_startino; + ig1->xi_alloccount = ig->xi_alloccount; + ig1->xi_allocmask = ig->xi_allocmask; +} |